CtrlK
BlogDocsLog inGet started
Tessl Logo

find-personalization-angles

Use when deriving evidence-backed outreach angles for a supplied contact from verified account context and dated research.

60

Quality

68%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/find-personalization-angles/SKILL.md
SKILL.md
Quality
Evals
Security

Find Personalization Angles

Purpose

Produce traceable personalization angles through fact-to-relevance-to-angle chains. Distinguish account context from personal knowledge.

Required inputs

  • Supplied contact identity, independently verified eligibility, and evidence
  • Supplied opt-out and stop-contact status
  • Supplied account record
  • Dated source excerpts with source type and citation
  • Approved offer facts and intended message goal
  • Observation window and evidence cutoff

Mark missing gate inputs unverified. Do not import out-of-scope evidence.

Workflow

  1. Check eligibility and opt-out status first. For any opt-out, stop request, or INELIGIBLE result, return STOP with no angles or copy. If either status is missing or unverified, return NEEDS VERIFICATION with no angles or copy.
  2. Inventory verified facts with subject, source type, date, citation, and freshness.
  3. Label each fact Account context or Personal knowledge. Company activity remains account context unless direct evidence ties the person to it.
  4. Map each fact to offer relevance as Inference without inventing pain, urgency, intent, or ownership.
  5. Draft an account-level angle. Never turn company activity into the person's action or belief.
  6. Rate evidence Strong, Moderate, or Weak; reject title-only, generic, stale, or familiar-sounding angles.
  7. Prefer fewer strong angles to filling the quota. Keep copy review-only.

Output format

# Personalization angles
- Status: READY, STOP, or NEEDS VERIFICATION
- Gate reason: [eligibility and opt-out evidence]
## Evidence boundary
- Account context: [verified facts]
- Personal knowledge: [verified facts or None supplied]
## Angles
### [angle name] — [Strong/Moderate/Weak]
- Fact: [claim, source type, date, citation]
- Relevance: [labeled inference]
- Angle: [review-only draft]
- Uncertainty: [limit]
## Rejected angles
- [generic or unsupported angle — reason]

Guardrails

  • Do not claim a person authored, led, owns, reviews, believes, needs, or prioritizes an account initiative without direct evidence. Do not use you or your team to attach company activity to that person.
  • Never draft an angle or copy after STOP or NEEDS VERIFICATION.
  • Reject generic praise, fake familiarity, fabricated common ground, inferred urgency, unsupported pain, and manipulative personalization.
  • Do not invent citations, identities, contactability, credentials, interests, relationships, behavior, intent, outcomes, or missing facts.
  • Keep facts, relevance inferences, and missing information explicit.
  • Require dated citations, source type, observation window, and evidence cutoff for researched claims.
  • Exclude protected-trait targeting and unnecessary sensitive personal data.
  • Preserve opt-outs and disqualifications.
  • Do not browse without authorization, scrape, contact people, mutate systems, or claim a draft was sent.

Quality check

Confirm status is first; opt-outs and ineligibility stop with no copy; missing gate evidence needs verification with no copy; and any ready angle has a cited fact-to-relevance chain without unsupported personal knowledge.

Example invocation

Use find-personalization-angles for Avery Lin with supplied contact, ELIGIBLE, and no opt-out or stop-contact request supplied statuses, Meridian account, and 2026-06-12 evidence. Return up to three cited chains.

Repository
llaskin/AI-SDR-Skill-Pack
Last updated
First committed

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.